On 3/10/25 15:12, Cheick TRAORE wrote:
> 
> On 3/10/25 13:35, Patrice CHOTARD wrote:
>>
>> On 3/10/25 11:00, Patrice CHOTARD wrote:
>>>
>>> On 3/6/25 15:13, Patrice CHOTARD wrote:
>>>>
>>>> On 3/6/25 11:56, Cheick Traore wrote:
>>>>> It was possible to provide a duty_ns greater than period_ns to
>>>>> "pwm config" command. The framework must check the values before
>>>>> providing them to drivers.
>>>>>
>>>>> Signed-off-by: Cheick Traore<cheick.tra...@foss.st.com>
>>>>> ---
>>>>>
>>>>>   drivers/pwm/pwm-uclass.c | 3 +++
>>>>>   1 file changed, 3 insertions(+)
>>>>>
>>>>> diff --git a/drivers/pwm/pwm-uclass.c b/drivers/pwm/pwm-uclass.c
>>>>> index 6543db1d623..b4491f7dcd4 100644
>>>>> --- a/drivers/pwm/pwm-uclass.c
>>>>> +++ b/drivers/pwm/pwm-uclass.c
>>>>> @@ -27,6 +27,9 @@ int pwm_set_config(struct udevice *dev, uint channel, 
>>>>> uint period_ns,
>>>>>       if (!ops->set_config)
>>>>>           return -ENOSYS;
>>>>>   +    if (duty_ns > period_ns)
>>>>> +        return -EINVAL;
>>>>> +
>>>>>       return ops->set_config(dev, channel, period_ns, duty_ns);
>>>>>   }
>>>>>   
>>>> Reviewed-by: Patrice Chotard<patrice.chot...@foss.st.com>
>>>>
>>>> Thanks
>>>> Patrice
>>> Applied to u-boot-stm32/next
>>>
>>> Thanks
>>> Patrice
>>
>> Hi Cheick
>>
>> Unfortunately, this patch is causing U-Boot CI test failed:
>> seehttps://source.denx.de/u-boot/custodians/u-boot-stm/-/jobs/1054426
>>
>> More precisely ut_dm_dm_test_cros_ec_pwm, see test/dm/cros_ec_pwm.c
>>
>> Either update test/dm/cros_ec_pwm.c or another solution is simply to
>> clamp duty_ns to period_ns as following ?
>>
>>   +    if (duty_ns > period_ns)
>>   +        duty_ns = period_ns;
>>   +
>>
>> Patrice
>>
>>
>>> _______________________________________________
>>> Uboot-stm32 mailing list
>>> uboot-st...@st-md-mailman.stormreply.com
>>> https://st-md-mailman.stormreply.com/mailman/listinfo/uboot-stm32
> 
> Hi Patrice,
> 
> It seems like this patch should finally be reverted.
> 
> When the duty cycle exceeds the period, some drivers already clamp the value 
> of the duty cycle to the maximum value or do not apply the duty cycle.
> 
> So, I will check the duty cycle in|drivers/pwm/pwm-stm32.c|and 
> return|-EINVAL|when it exceeds the period.
> 
> Thanks,
> 
> Cheick

ok, i will remove your PWM series from the STM32_U-boot_custodian/next branch.

Thanks
Patrice

> 

Reply via email to